def plotXPropTempDensity(fileh5, time=0, Z=4, Y=6):
markers = ['v-c', '^-y', '<-r', '>-m', '*-b', 'd-g', 'p-b', '1-r', '2-m']
pl = []
pl_name = []
D = gkcData.getDomain(fileh5)
# we iterate of all species
for s in range(len(fileh.root.Phasespace.Data[:, 0, 0, 0, 0, 0, 0])):
species_name = fileh.root.Species.cols.Name[s]
species_name = "species"
x = D['X']
pl.append(plot(x, Density(time, s)[Z, Y, :], markers[2 * s]))
pl_name.append("Density (" + species_name + ")")
xlabel("Position [x]")
ylabel("Density")
# set scaling for density
d = Density(time)[Z, Y, :]
if ((max(d) - min(d)) < 0.2): ylim(min(d) - 0.1, max(d) + 0.1)
twinx()
pl.append(plot(x, Temperature(time, s)[Z, Y, :], markers[2 * s + 1]))
pl_name.append("Temperature (" + species_name + ")")
ylabel("Temperature")
title("Density/Temperature profile at Time Step = " + str(time))
leg = pylab.legend(pl, pl_name, loc='lower center')
leg.draw_frame(0)
def plotTurbulenceTime(fileh5, dir='Y', pos=(-2,-1), doFit='False', posT=(1,-1)):
"""
Plots time evolution of mode power.
Optional keyword arguments:
Keyword Description
=============== ==============================================
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*modes* List of modes (default plot modes).
e.g. modes = [1,4,5] - to plot all modes
modes = range(Nky)[::2] - to plot every second mode
*field* 'phi' electric potential
'A' parallel magnetic vector potential
'B' parallel magnetic field
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*doCFL* clear previous figure
*label* 'ky' or 'm'
*offset* Offset due to zeroset to 2 .
"""
D = gkcData.getDomain(fileh5)
data = fileh5.root.Analysis.PowerSpectrum.Y[1:,:]
T = gkcData.getTime(fileh5.root.Analysis.PowerSpectrum.Time)[:,1]
timeEvolution = []
for step in range(len(data[0,:])):
timeEvolution.append(data[:,step]/sum(data[:,step]))
contourf(np.log10(D['ky']), T,timeEvolution, 250, locator=ticker.LogLocator(), cmap=cm.jet)
colorbar()
def plotXPropTempDensity(fileh5, time=0, Z=4, Y=6):
markers = ['v-c', '^-y', '<-r', '>-m', '*-b', 'd-g', 'p-b', '1-r', '2-m']
pl = []
pl_name = []
D = gkcData.getDomain(fileh5)
# we iterate of all species
for s in range(len(fileh.root.Phasespace.Data[:,0,0,0,0,0,0])):
species_name = fileh.root.Species.cols.Name[s]
species_name = "species"
x = D['X']
pl.append(plot(x, Density(time, s)[Z,Y,:], markers[2*s]))
pl_name.append("Density (" + species_name + ")")
xlabel("Position [x]")
ylabel("Density")
# set scaling for density
d = Density(time)[Z,Y,:]
if((max(d) - min(d)) < 0.2): ylim(min(d)-0.1, max(d)+0.1)
twinx()
pl.append(plot(x, Temperature(time, s)[Z,Y,:],markers[2*s+1]))
pl_name.append("Temperature (" + species_name + ")")
ylabel("Temperature")
title("Density/Temperature profile at Time Step = " + str(time))
leg = pylab.legend(pl, pl_name, loc='lower center')
leg.draw_frame(0)
def plotEigenfunctions(fileh5, mode=1, n_max=5, **kwargs):
"""
Plot the n largerst eigenfunctions
Optional keyword arguments:
Keyword Description
=============== ==============================================
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*offset* Offset due to zeroset to 2 .
"""
import gkcStyle
import gkcAnalysis
D = gkcData.getDomain(fileh5)
try:
eigv = fileh5.root.Eigenvalue.EigenValues.cols.Eigenvalue[:]
except:
print "Problems openning eigenvalues. Not included ?"
return
# Sort eigenvalues (note automatically sorts real values first)
eigv = fileh5.root.Eigenvalue.EigenValues.cols.Eigenvalue[:]
# absteigend sort
idx_sort = np.argsort(eigv)[::-1]
pylab.subplot(121)
plotEigenvalues(fileh5, **kwargs)
# Plot points
for n in range(n_max):
w = eigv[idx_sort[n]]
pylab.plot(np.real(w),
np.imag(w),
'o',
markersize=7.5,
color=gkcStyle.markers_C[n])
pylab.subplot(122)
labels = []
for n in range(n_max):
w = eigv[idx_sort[n]]
gkcAnalysis.plotModeStructure(fileh5=fileh5,
fied='phi',
mode=mode,
frame=idx_sort[n],
part='a',
m=n)
labels.append("%.3f+%.3f i" % (np.real(w), np.imag(w)))
pylab.legend(labels).draw_frame(0)
def plotModeStructure(fileh5, mode, part = "r", **kwargs):
"""
Plots time evolution of mode power.
Optional keyword arguments:
Keyword Description
=============== ==============================================
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*modes* List of modes (default plot modes).
e.g. modes = [1,4,5] - to plot all modes
modes = range(Nky)[::2] - to plot every second mode
*field* 'phi' electric potential
'A' parallel magnetic vector potential
'B' parallel magnetic field
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*doCFL* clear previous figure
*label* 'ky' or 'm'
*offset* Offset due to zeroset to 2 .
"""
D = gkcData.getDomain(fileh5)
doCFL = kwargs.pop('doCFL', True)
field = kwargs.pop('field', 'phi')
Z = kwargs.pop('Z', 0)
frame = kwargs.pop('frame', -1)
label = kwargs.pop('label', "")
m = kwargs.pop('m', 0)
if field == 'phi' : data_X = fileh5.root.Visualization.Phi[Z,mode,:,frame]
elif field == 'A' : n_field = 1
elif field == 'B' : n_field = 2
else : raise TypeError("Wrong argument for field : " + str(field))
# Normaliza
if part == "a" : data_X = abs(data_X) / np.sum(abs(data_X))
elif part == "r" : data_X = np.real(data_X) / np.sum(abs(np.real(data_X)))
elif part == "i" : data_X = np.imag(data_X) / np.sum(abs(np.imag(data_X)))
else : raise TypeError("Wrong argument for part : " + str(part))
pylab.plot(D['X'], data_X, gkcStyle.markers_C[m], label=label)
if field == "phi" : pylab.ylabel("Mode Power $|\\phi|^2$")
elif field == "A" : pylab.ylabel("Mode Power $|A_\\parallel|^2$")
elif field == "B" : pylab.ylabel("Mode Power $|B_\\parallel|^2$")
else : raise TypeError("Wrong argument for field : " + str(field))
pylab.xlim((min(D['X']), max(D['X'])))
pylab.xlabel("X")
def plotEigenfunctions(fileh5, mode=1, n_max=5, **kwargs):
"""
Plot the n largerst eigenfunctions
Optional keyword arguments:
Keyword Description
=============== ==============================================
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*offset* Offset due to zeroset to 2 .
"""
import gkcStyle
import gkcAnalysis
D = gkcData.getDomain(fileh5)
try:
eigv = fileh5.root.Eigenvalue.EigenValues.cols.Eigenvalue[:]
except:
print "Problems openning eigenvalues. Not included ?"
return
# Sort eigenvalues (note automatically sorts real values first)
eigv = fileh5.root.Eigenvalue.EigenValues.cols.Eigenvalue[:]
# absteigend sort
idx_sort = np.argsort(eigv)[::-1]
pylab.subplot(121)
plotEigenvalues(fileh5, **kwargs)
# Plot points
for n in range(n_max):
w = eigv[idx_sort[n]]
pylab.plot(np.real(w), np.imag(w), 'o', markersize=7.5, color=gkcStyle.markers_C[n])
pylab.subplot(122)
labels = []
for n in range(n_max):
w = eigv[idx_sort[n]]
gkcAnalysis.plotModeStructure(fileh5=fileh5, fied='phi', mode=mode, frame=idx_sort[n], part='a', m=n)
labels.append("%.3f+%.3f i" % (np.real(w), np.imag(w)))
pylab.legend(labels).draw_frame(0)
def plotContour(fileh5, var="2DPhi", **kwargs):
"""
Plots Scalar Data (2D) in XY coordinates
Optional keyword arguments:
Keyword Description
=============== ==============================================
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*modes* List of modes (default plot modes).
e.g. modes = [1,4,5] - to plot all modes
modes = range(Nky)[::2] - to plot every second mode
*field* 'phi' electric potential
'A' parallel magnetic vector potential
'B' parallel magnetic field
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*doCFL* clear previous figure
*label* 'ky' or 'm'
*offset* Offset due to zeroset to 2 .
"""
D = gkcData.getDomain(fileh5)
#norm = kwargs.pop('Normalize', True)
Z = kwargs.pop('Z', 0)
modes = kwargs.pop('modes' , range(D['Nky']))
doCFL = kwargs.pop('doCFL' , True)
#interpolation = kwargs.pop('interpolation' , 'bilinear')
printTitle = kwargs.pop('printTitle' , True)
#orientation = kwargs.pop('orientation' , 'horizontal')
frame = kwargs.pop('frame' , -1)
D, T, data = gkcData.getData(var, fileh5, Z, frame, species=0)
X, Y, data = gkcData.getRealFromXky(fileh5, data, modes)
gkcStyle.plotContourWithColorbar(X,Y, data, **kwargs)
pylab.xlabel(kwargs.pop('xlabel' , 'X'))
pylab.ylabel(kwargs.pop('ylabel' , 'Y'))
if printTitle == True : pylab.title("TimeStep : %i Time : %.3f " % (T[0], T[1]))
def plotContour(fileh5, var="2DPhi", **kwargs):
"""
Plots Scalar Data (2D) in XY coordinates
Optional keyword arguments:
Keyword Description
=============== ==============================================
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*modes* List of modes (default plot modes).
e.g. modes = [1,4,5] - to plot all modes
modes = range(Nky)[::2] - to plot every second mode
*field* 'phi' electric potential
'A' parallel magnetic vector potential
'B' parallel magnetic field
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*doCFL* clear previous figure
*label* 'ky' or 'm'
*offset* Offset due to zeroset to 2 .
"""
D = gkcData.getDomain(fileh5)
#norm = kwargs.pop('Normalize', True)
Z = kwargs.pop('Z', 0)
modes = kwargs.pop('modes', range(D['Nky']))
doCFL = kwargs.pop('doCFL', True)
#interpolation = kwargs.pop('interpolation' , 'bilinear')
printTitle = kwargs.pop('printTitle', True)
#orientation = kwargs.pop('orientation' , 'horizontal')
frame = kwargs.pop('frame', -1)
D, T, data = gkcData.getData(var, fileh5, Z, frame, species=0)
X, Y, data = gkcData.getRealFromXky(fileh5, data, modes)
gkcStyle.plotContourWithColorbar(X, Y, data, **kwargs)
pylab.xlabel(kwargs.pop('xlabel', 'X'))
pylab.ylabel(kwargs.pop('ylabel', 'Y'))
if printTitle == True:
pylab.title("TimeStep : %i Time : %.3f " % (T[0], T[1]))
def plotEigenvalues(fileh5, **kwargs):
"""
Plot the eigenvalues of the phase space function
Optional keyword arguments:
Keyword Description
=============== ==============================================
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*offset* Offset due to zeroset to 2 .
"""
import gkcStyle
D = gkcData.getDomain(fileh5)
try:
eigv = fileh5.root.Eigenvalue.EigenValues.cols.Eigenvalue[:]
except:
print "Problems openning eigenvalues. Not included ?"
return
eigv_r = np.real(eigv)
eigv_i = np.imag(eigv)
pylab.plot(eigv_r, eigv_i, '.')
gkcStyle.plotZeroLine(1.1 * min(eigv_r),
1.1 * max(eigv_r),
direction='horizontal',
color="#666666",
lw=0.8)
gkcStyle.plotZeroLine(1.1 * min(eigv_i),
1.1 * max(eigv_i),
direction='vertical',
color="#666666",
lw=0.8)
pylab.xlim((1.1 * min(eigv_r), 1.1 * max(eigv_r)))
pylab.ylim((1.1 * min(eigv_i), 1.1 * max(eigv_i)))
pylab.xlabel("$\\omega_r$")
pylab.ylabel("$\\omega_i$")
def plotEigenvalues(fileh5, **kwargs):
"""
Plot the eigenvalues of the phase space function
Optional keyword arguments:
Keyword Description
=============== ==============================================
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*offset* Offset due to zeroset to 2 .
"""
import gkcStyle
D = gkcData.getDomain(fileh5)
try:
eigv = fileh5.root.Eigenvalue.EigenValues.cols.Eigenvalue[:]
except:
print "Problems openning eigenvalues. Not included ?"
return
eigv_r = np.real(eigv)
eigv_i = np.imag(eigv)
pylab.plot(eigv_r, eigv_i, '.')
gkcStyle.plotZeroLine(1.1*min(eigv_r), 1.1*max(eigv_r), direction='horizontal', color="#666666", lw=0.8)
gkcStyle.plotZeroLine(1.1*min(eigv_i), 1.1*max(eigv_i), direction='vertical' , color="#666666", lw=0.8)
pylab.xlim((1.1*min(eigv_r), 1.1*max(eigv_r)))
pylab.ylim((1.1*min(eigv_i), 1.1*max(eigv_i)))
pylab.xlabel("$\\omega_r$")
pylab.ylabel("$\\omega_i$")
def plotTurbulenceTime(fileh5,
dir='Y',
pos=(-2, -1),
doFit='False',
posT=(1, -1)):
"""
Plots time evolution of mode power.
Optional keyword arguments:
Keyword Description
=============== ==============================================
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*modes* List of modes (default plot modes).
e.g. modes = [1,4,5] - to plot all modes
modes = range(Nky)[::2] - to plot every second mode
*field* 'phi' electric potential
'A' parallel magnetic vector potential
'B' parallel magnetic field
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*doCFL* clear previous figure
*label* 'ky' or 'm'
*offset* Offset due to zeroset to 2 .
"""
D = gkcData.getDomain(fileh5)
data = fileh5.root.Analysis.PowerSpectrum.Y[1:, :]
T = gkcData.getTime(fileh5.root.Analysis.PowerSpectrum.Time)[:, 1]
timeEvolution = []
for step in range(len(data[0, :])):
timeEvolution.append(data[:, step] / sum(data[:, step]))
contourf(np.log10(D['ky']),
T,
timeEvolution,
250,
locator=ticker.LogLocator(),
cmap=cm.jet)
colorbar()
def plotScalarDataTimeEvolution(fileh5, var="KinPhi", **kwargs):
"""
Plots time evolution of mode power.
Optional keyword arguments:
Keyword Description
=============== ==============================================
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*modes* List of modes (default plot modes).
e.g. modes = [1,4,5] - to plot all modes
modes = range(Nky)[::2] - to plot every second mode
*field* 'phi' electric potential
'A' parallel magnetic vector potential
'B' parallel magnetic field
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*doCFL* clear previous figure
*label* 'ky' or 'm'
*offset* Offset due to zeroset to 2 .
"""
D = gkcData.getDomain(fileh5)
log = kwargs.pop('log', True)
leg_loc = kwargs.pop('leg_loc', 'best')
leg_ncol = kwargs.pop('leg_ncol', 1)
if (log == 1):
plotf = pylab.semilogy
af = abs
else:
plotf = pylab.plot
af = lambda x: x
T = fileh5.root.Analysis.scalarValues.cols.Time[5:]
n_sp = len(fileh5.root.Species.cols.Name[1:])
# Plot Field Energy"
if var == "phi":
plotf(T,
af(fileh5.root.Analysis.scalarValues.cols.phiEnergy[3:]),
label="Field Energy")
pylab.ylabel("Field Energy")
if var == "kinEnergy":
for s in range(n_sp):
plotf(T,
af(fileh5.root.Analysis.scalarValues.cols.KineticEnergy[3:]
[:, s]),
label="Kinetic Energy (" + fileh5.root.Species.cols.Name[s] +
")")
#plotf(T, af(fileh5.root.Analysis.scalarValues.cols.phiEnergy[3:]+1.*(np.sum(fileh5.root.Analysis.scalarValues.cols.KineticEnergy[3:], axis=1))) , label="Total Energy")
pylab.ylabel("Energy")
if var == "KinPhi":
pylab.plot(T,
fileh5.root.Analysis.scalarValues.cols.phiEnergy[5:],
label="Field Energy")
E_kin_total = fileh5.root.Analysis.scalarValues.cols.phiEnergy[5:]
for s in range(n_sp):
E_kin = fileh5.root.Analysis.scalarValues.cols.KineticEnergy[
5:][:,
s] - fileh5.root.Analysis.scalarValues.cols.KineticEnergy[
5][s]
E_kin = -abs(E_kin)
E_kin_total = E_kin_total + E_kin
pylab.plot(T,
E_kin,
label="Kinetic Energy (" +
fileh5.root.Species.cols.Name[s] + ")")
gkcStyle.plotZeroLine(min(T), max(T))
pylab.plot(T, -abs(E_kin_total) / 100., label="Total Energy")
pylab.yscale('symlog', linthreshy=1.e-7)
# Plot Toal nergy
pylab.ylabel("Energy")
if var == "heatFlux":
for s in sp:
for s in range(n_sp):
plot(T,
fileh5.root.Analysis.scalarValues.cols.HeatFlux[3:][:, s],
label=fileh5.root.Species.cols.Name[s])
pylab.ylabel("Heat Flux")
if var == "Charge":
charge = np.zeros(
len(fileh5.root.Analysis.scalarValues.cols.ParticleNumber[3:][:,
1]))
for s in range(n_sp):
dn = (
fileh5.root.Analysis.scalarValues.cols.ParticleNumber[3:][:, s]
- fileh5.root.Analysis.scalarValues.cols.ParticleNumber[3][s])
n = fileh5.root.Analysis.scalarValues.cols.ParticleNumber[3:][:, s]
q = fileh5.root.Species.cols.Charge[s]
charge = charge + q * n
pylab.plot(fileh5.root.Analysis.scalarValues.cols.Time[3:],
q * n,
label=fileh5.root.Species.cols.Name[s])
pylab.plot(fileh5.root.Analysis.scalarValues.cols.Time[3:],
q * dn,
label=fileh5.root.Species.cols.Name[s])
if D['Ns'] > 1:
plotf(fileh5.root.Analysis.scalarValues.cols.Time[3:],
charge,
label="Charge")
gkcStyle.plotZeroLine(min(T), max(T))
pylab.yscale('symlog', linthreshy=1.e-7)
pylab.ylabel("Charge")
if var == "particleFlux":
for s in range(n_sp):
plotf(T,
fileh5.root.Analysis.scalarValues.cols.ParticleFlux[3:][:,
s],
label=fileh5.root.Species.cols.Name[s])
pylab.ylabel("Particle Flux")
pylab.xlim((min(T), max(T)))
leg = pylab.legend(loc=leg_loc, ncol=leg_ncol).draw_frame(0)
pylab.xlabel("Time")
def plotTurbulenceSpectra(fileh5, dir='Y', start=1, end=-1, posT=(1,-1), field=0, **kwargs):
"""
Plots time evolution of mode power.
Optional keyword arguments:
Keyword Description
=============== ==============================================
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*modes* List of modes (default plot modes).
e.g. modes = [1,4,5] - to plot all modes
modes = range(Nky)[::2] - to plot every second mode
*field* 'phi' electric potential
'A' parallel magnetic vector potential
'B' parallel magnetic field
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*doCFL* clear previous figure
*label* 'ky' or 'm'
*offset* Offset due to zeroset to 2 .
"""
import scipy.optimize
D = gkcData.getDomain(fileh5)
doCFL = kwargs.pop('doCFL', True)
doFit = kwargs.pop('doFit', False)
doZF = kwargs.pop('doZF' , True)
m = kwargs.pop('m' , 0)
if(dir == 'X'):
data = sum(fileh5.root.Analysis.PowerSpectrum.X[field,0:,start:end], axis=1)/abs(end-start)
loglog(D['kx'], data/sum(data), '.-')
xlabel("$k_x$")
elif(dir == 'Y'):
data = np.mean(fileh5.root.Analysis.PowerSpectrum.Y[field, 0:,start:end], axis=1)
mypl = pylab.loglog(D['ky'][1:-1], data[1:-1], gkcStyle.markers_D[m] + '-', color=gkcStyle.markers_C[m])
# Plot Zonal flow seperately
pylab.loglog(0.9*D['ky'][1], data[0], gkcStyle.markers_D[m], markersize=8., color=gkcStyle.markers_C[m])
pylab.xlim((0.75 * D['ky'][1], 1.25*D['ky'][-2]))
pylab.xlabel("$k_y$")
# Draw vertical line
if doZF == 0:
min_x = min(data)
max_x = max(data)
pylab.loglog(np.linspace(0.9*D['ky'][1], 0.9*D['ky'][1], 201), np.logspace(np.log10(0.8*min_x), np.log10(1.2*max_x), 201), "-", linewidth=1.5, color="#666666")
#pylab.text(D['ky'][1], 1.05*np.sqrt(max_x), "Zonal Flow", weight='bold', color="#666666", rotation='vertical')
#pylab.text(0.95*D['ky'][1], 5*min_x, "Zonal Flow", weight='bold', color="#666666", rotation='vertical')
elif(dir == 'Z'):
data = sum(fileh5.root.Analysis.PowerSpectrum.Z[field, 0:,start:end], axis=1)/abs(start-end)
pylab.loglog(D['kp'], data[1:], '.-')
pylab.loglog(D['kp'][0], data[0], 'o')
xlabel("$k_z$")
#pylab.ylabel("$|\\phi_k(k_y)|^2$")
"""
if doFit == True :
pos_a = posT[0]
pos_b = posT[1]
fitfunc = lambda p, x: p[0]*x + p[1] # Target function
errfunc = lambda p, x, y: fitfunc(p, x) - y # Distance to the target function
p0 = [1.0, 1.0, 1.0] # Initial guess for the parameters
#p1, success = scipy.optimize.leastsq(errfunc, p0[:], args=(np.log10(D['ky'][pos_a:pos_b]), np.log10(data[pos_a:pos_b])))
p1, success = scipy.optimize.leastsq(errfunc, p0[:], args=(D['ky'][pos_a:pos_b], data[pos_a:pos_b]))
# C'mon baby wanna see you again
print p1
pylab.loglog(D['ky'][pos_a:pos_b], p1[1]*(D['ky'][pos_a:pos_b]/D['ky'][pos_a])**p1[0],'k-', linewidth=6.)
#text(D['ky'][5], data[5], "$\\propto k_y^{%2.f}$" % p1[0], ha="center", family=font, size=14)
pos_m = (pos_a + pos_b)/2 + 1
pylab.text(D['ky'][pos_m], data[pos_m/3], "$\\propto k_y^{%.1f}$" % p1[0], ha="center", size=14)
#pylab.text(np.log(D['ky'][5]), 1., "$\\propto k_y^{%.1f}$" % p1[0], ha="center", size=14)
#pylab.text(np.log(D['ky'][5]), np.log(data[5]), "$\\propto k_y^{%2.1f}$" % p1[0], ha="center", size=14)
"""
return mypl
def plotScalarDataTimeEvolution(fileh5, var = "KinPhi", **kwargs):
"""
Plots time evolution of mode power.
Optional keyword arguments:
Keyword Description
=============== ==============================================
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*modes* List of modes (default plot modes).
e.g. modes = [1,4,5] - to plot all modes
modes = range(Nky)[::2] - to plot every second mode
*field* 'phi' electric potential
'A' parallel magnetic vector potential
'B' parallel magnetic field
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*doCFL* clear previous figure
*label* 'ky' or 'm'
*offset* Offset due to zeroset to 2 .
"""
D = gkcData.getDomain(fileh5)
log = kwargs.pop('log', True)
leg_loc = kwargs.pop('leg_loc' , 'best')
leg_ncol = kwargs.pop('leg_ncol', 1)
if(log == 1) :
plotf = pylab.semilogy
af = abs
else :
plotf = pylab.plot
af = lambda x : x
T = fileh5.root.Analysis.scalarValues.cols.Time[5:]
n_sp = len(fileh5.root.Species.cols.Name[1:])
# Plot Field Energy"
if var == "phi":
plotf(T, af(fileh5.root.Analysis.scalarValues.cols.phiEnergy[3:]), label="Field Energy")
pylab.ylabel("Field Energy")
if var == "kinEnergy":
for s in range(n_sp):
plotf(T, af(fileh5.root.Analysis.scalarValues.cols.KineticEnergy[3:][:,s]) , label="Kinetic Energy (" + fileh5.root.Species.cols.Name[s]+ ")" )
#plotf(T, af(fileh5.root.Analysis.scalarValues.cols.phiEnergy[3:]+1.*(np.sum(fileh5.root.Analysis.scalarValues.cols.KineticEnergy[3:], axis=1))) , label="Total Energy")
pylab.ylabel("Energy")
if var == "KinPhi":
pylab.plot(T, fileh5.root.Analysis.scalarValues.cols.phiEnergy[5:], label="Field Energy")
E_kin_total = fileh5.root.Analysis.scalarValues.cols.phiEnergy[5:]
for s in range(n_sp):
E_kin = fileh5.root.Analysis.scalarValues.cols.KineticEnergy[5:][:,s]-fileh5.root.Analysis.scalarValues.cols.KineticEnergy[5][s]
E_kin = -abs(E_kin)
E_kin_total = E_kin_total + E_kin
pylab.plot(T, E_kin , label="Kinetic Energy (" + fileh5.root.Species.cols.Name[s]+ ")" )
gkcStyle.plotZeroLine(min(T), max(T))
pylab.plot(T, -abs(E_kin_total)/100. , label="Total Energy")
pylab.yscale('symlog', linthreshy=1.e-7)
# Plot Toal nergy
pylab.ylabel("Energy")
if var == "heatFlux":
for s in sp:
for s in range(n_sp):
plot(T, fileh5.root.Analysis.scalarValues.cols.HeatFlux[3:][:,s], label=fileh5.root.Species.cols.Name[s])
pylab.ylabel("Heat Flux")
if var == "Charge":
charge = np.zeros(len(fileh5.root.Analysis.scalarValues.cols.ParticleNumber[3:][:,1]))
for s in range(n_sp):
dn = (fileh5.root.Analysis.scalarValues.cols.ParticleNumber[3:][:,s] - fileh5.root.Analysis.scalarValues.cols.ParticleNumber[3][s])
n = fileh5.root.Analysis.scalarValues.cols.ParticleNumber[3:][:,s]
q = fileh5.root.Species.cols.Charge[s]
charge = charge + q * n
pylab.plot(fileh5.root.Analysis.scalarValues.cols.Time[3:], q*n , label=fileh5.root.Species.cols.Name[s])
pylab.plot(fileh5.root.Analysis.scalarValues.cols.Time[3:], q*dn , label=fileh5.root.Species.cols.Name[s])
if D['Ns'] > 1:
plotf(fileh5.root.Analysis.scalarValues.cols.Time[3:], charge , label="Charge")
gkcStyle.plotZeroLine(min(T), max(T))
pylab.yscale('symlog', linthreshy=1.e-7)
pylab.ylabel("Charge")
if var == "particleFlux":
for s in range(n_sp):
plotf(T, fileh5.root.Analysis.scalarValues.cols.ParticleFlux[3:][:,s], label=fileh5.root.Species.cols.Name[s])
pylab.ylabel("Particle Flux")
pylab.xlim((min(T), max(T)))
leg = pylab.legend(loc=leg_loc, ncol=leg_ncol).draw_frame(0)
pylab.xlabel("Time")
def plotAveragedFluxes(fileh5, var='H', **kwargs):
"""
Plots time evolution of mode power.
Optional keyword arguments:
Keyword Description
=============== ==============================================
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*modes* List of modes (default plot modes).
e.g. modes = [1,4,5] - to plot all modes
modes = range(Nky)[::2] - to plot every second mode
*field* 'phi' electric potential
'A' parallel magnetic vector potential
'B' parallel magnetic field
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*doCFL* clear previous figure
*label* 'ky' or 'm'
*offset* Offset due to zeroset to 2 .
"""
import gkcData
import gkcStyle
import pylab
import numpy as np
D = gkcData.getDomain(fileh5)
doCFL = kwargs.pop('doCFL', True)
start = kwargs.pop('start', 1)
stop = kwargs.pop('stop', -1)
scale = kwargs.pop('scale', 'ky')
field = kwargs.pop('field', 'phi')
if doCFL == True: pylab.clf()
if field == 'phi': n_field = 0
elif field == 'A': n_field = 1
elif field == 'B': n_field = 2
else: raise TypeError("Wrong argument for field : " + str(field))
T = gkcData.getTime(fileh5.root.Analysis.PowerSpectrum.Time)[:, 1]
# Averaged over Z and start stop
if var == 'P':
data = np.mean(fileh5.root.Analysis.Flux.Density[n_field, :, :,
start:stop],
axis=2)
ylabel = "Particle Flux $k_y \\Gamma/\\Gamma_{gB}\,(ky)$"
elif var == 'H':
data = np.mean(fileh5.root.Analysis.Flux.Heat[n_field, :, :,
start:stop],
axis=2)
#ylabel = "Heat Flux $Q/Q_\\textrm{gB}(k_y)$"
ylabel = "Heat Flux $k_y Q/Q_{gB}\,(k_y)$"
else:
raise TypeError("No such variable")
if scale == 'ky': gor_ky = D['ky'][1:]
else: gor_ky = 1.
for s in range(D['Ns']):
species_name = fileh5.root.Species.cols.Name[s + 1]
pylab.semilogx(D['ky'][1:],
gor_ky * data[1:, s],
gkcStyle.markers_D[s] + '-',
label=species_name,
color=gkcStyle.markers_C[s],
markersize=7.)
pylab.xlabel("$k_y \\rho_i$")
pylab.ylabel(ylabel)
pylab.legend(ncol=2).draw_frame(0)
pylab.xlim((0.8 * D['ky'][1], 1.2 * D['ky'][-1]))
def plotTurbulenceSpectra(fileh5,
dir='Y',
start=1,
end=-1,
posT=(1, -1),
field=0,
**kwargs):
"""
Plots time evolution of mode power.
Optional keyword arguments:
Keyword Description
=============== ==============================================
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*modes* List of modes (default plot modes).
e.g. modes = [1,4,5] - to plot all modes
modes = range(Nky)[::2] - to plot every second mode
*field* 'phi' electric potential
'A' parallel magnetic vector potential
'B' parallel magnetic field
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*doCFL* clear previous figure
*label* 'ky' or 'm'
*offset* Offset due to zeroset to 2 .
"""
import scipy.optimize
D = gkcData.getDomain(fileh5)
doCFL = kwargs.pop('doCFL', True)
doFit = kwargs.pop('doFit', False)
doZF = kwargs.pop('doZF', True)
m = kwargs.pop('m', 0)
if (dir == 'X'):
data = sum(fileh5.root.Analysis.PowerSpectrum.X[field, 0:, start:end],
axis=1) / abs(end - start)
loglog(D['kx'], data / sum(data), '.-')
xlabel("$k_x$")
elif (dir == 'Y'):
data = np.mean(fileh5.root.Analysis.PowerSpectrum.Y[field, 0:,
start:end],
axis=1)
mypl = pylab.loglog(D['ky'][1:-1],
data[1:-1],
gkcStyle.markers_D[m] + '-',
color=gkcStyle.markers_C[m])
# Plot Zonal flow seperately
pylab.loglog(0.9 * D['ky'][1],
data[0],
gkcStyle.markers_D[m],
markersize=8.,
color=gkcStyle.markers_C[m])
pylab.xlim((0.75 * D['ky'][1], 1.25 * D['ky'][-2]))
pylab.xlabel("$k_y$")
# Draw vertical line
if doZF == 0:
min_x = min(data)
max_x = max(data)
pylab.loglog(np.linspace(0.9 * D['ky'][1], 0.9 * D['ky'][1], 201),
np.logspace(np.log10(0.8 * min_x),
np.log10(1.2 * max_x), 201),
"-",
linewidth=1.5,
color="#666666")
#pylab.text(D['ky'][1], 1.05*np.sqrt(max_x), "Zonal Flow", weight='bold', color="#666666", rotation='vertical')
#pylab.text(0.95*D['ky'][1], 5*min_x, "Zonal Flow", weight='bold', color="#666666", rotation='vertical')
elif (dir == 'Z'):
data = sum(fileh5.root.Analysis.PowerSpectrum.Z[field, 0:, start:end],
axis=1) / abs(start - end)
pylab.loglog(D['kp'], data[1:], '.-')
pylab.loglog(D['kp'][0], data[0], 'o')
xlabel("$k_z$")
#pylab.ylabel("$|\\phi_k(k_y)|^2$")
"""
if doFit == True :
pos_a = posT[0]
pos_b = posT[1]
fitfunc = lambda p, x: p[0]*x + p[1] # Target function
errfunc = lambda p, x, y: fitfunc(p, x) - y # Distance to the target function
p0 = [1.0, 1.0, 1.0] # Initial guess for the parameters
#p1, success = scipy.optimize.leastsq(errfunc, p0[:], args=(np.log10(D['ky'][pos_a:pos_b]), np.log10(data[pos_a:pos_b])))
p1, success = scipy.optimize.leastsq(errfunc, p0[:], args=(D['ky'][pos_a:pos_b], data[pos_a:pos_b]))
# C'mon baby wanna see you again
print p1
pylab.loglog(D['ky'][pos_a:pos_b], p1[1]*(D['ky'][pos_a:pos_b]/D['ky'][pos_a])**p1[0],'k-', linewidth=6.)
#text(D['ky'][5], data[5], "$\\propto k_y^{%2.f}$" % p1[0], ha="center", family=font, size=14)
pos_m = (pos_a + pos_b)/2 + 1
pylab.text(D['ky'][pos_m], data[pos_m/3], "$\\propto k_y^{%.1f}$" % p1[0], ha="center", size=14)
#pylab.text(np.log(D['ky'][5]), 1., "$\\propto k_y^{%.1f}$" % p1[0], ha="center", size=14)
#pylab.text(np.log(D['ky'][5]), np.log(data[5]), "$\\propto k_y^{%2.1f}$" % p1[0], ha="center", size=14)
"""
return mypl
def plotTimeEvolutionModePower(fileh5, **kwargs):
"""
Plots time evolution of mode power.
Optional keyword arguments:
Keyword Description
=============== ==============================================
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*modes* List of modes (default plot modes).
e.g. modes = [1,4,5] - to plot all modes
modes = range(Nky)[::2] - to plot every second mode
*field* 'phi' electric potential
'A' parallel magnetic vector potential
'B' parallel magnetic field
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*doCFL* clear previous figure
*label* 'ky' or 'm'
*offset* Offset due to zeroset to 2 .
"""
D = gkcData.getDomain(fileh5)
doCFL = kwargs.pop('doCFL', True)
doLog = kwargs.pop('doLog', True)
dir = kwargs.pop('dir', 'Y')
modes = kwargs.pop('modes', range(D['Nky']))
field = kwargs.pop('field', 'phi')
n_offset = kwargs.pop('offset', 3)
label = kwargs.pop('label', 'k')
leg_loc = kwargs.pop('loc', 'best')
ncol = kwargs.pop('ncol', 3)
showLegend = kwargs.pop('showLegend', True)
if doCFL == True: pylab.clf()
T = gkcData.getTime(fileh5.root.Analysis.PowerSpectrum.Time)[:, 1]
if field == 'phi': n_field = 0
elif field == 'A': n_field = 1
elif field == 'B': n_field = 2
else: raise TypeError("Wrong argument for field : " + str(field))
legend_list = []
if (dir == 'X'):
for n in modes:
pl = pylab.semilogy(
T[n_offset:],
fileh5.root.Analysis.PowerSpectrum.X[n_field, n, n_offset:].T)
if (label == 'm'): legend_list.append("n = %i" % n)
elif (label == 'k'): legend_list.append("kx = %.1f" % (D['kx'][n]))
else: print "Name Error"
elif (dir == 'Y'):
legend_list = []
for m in modes:
if (doLog):
pylab.semilogy(
T[n_offset:],
fileh5.root.Analysis.PowerSpectrum.Y[n_field, m,
n_offset:].T,
label='$k_y^{(%i)} = %.2f$' % (m, D['ky'][m]))
else:
pylab.plot(T[n_offset:],
fileh5.root.Analysis.PowerSpectrum.Y[n_field, m,
n_offset:].T,
label='$k_y^{(%i)} = %.2f$' % (m, D['ky'][m]))
else:
raise TypeError("Wrong argument for dir : " + str(dir))
#if showLegend == True:
leg = pylab.legend(loc=leg_loc, ncol=ncol, mode="expand").draw_frame(0)
pylab.xlabel("Time")
pylab.xlim((0., max(T)))
if field == "phi": pylab.ylabel("Mode Power $|\\phi|^2$")
elif field == "A": pylab.ylabel("Mode Power $|A_\\parallel|^2$")
elif field == "B": pylab.ylabel("Mode Power $|B_\\parallel|^2$")
else: raise TypeError("Wrong argument for field : " + str(field))
def plotTimeEvolutionModePhase(fileh5, **kwargs):
"""
Plots time evolution of mode power.
Optional keyword arguments:
Keyword Description
=============== ==============================================
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*modes* List of modes (default plot modes).
e.g. modes = [1,4,5] - to plot all modes
modes = range(Nky)[::2] - to plot every second mode
*field* 'phi' electric potential
'A' parallel magnetic vector potential
'B' parallel magnetic field
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*doCFL* clear previous figure
*label* 'ky' or 'm'
*offset* Offset due to zeroset to 2 .
"""
D = gkcData.getDomain(fileh5)
doCFL = kwargs.pop('doCFL', True)
leg_loc = kwargs.pop('loc', 'best')
dir = kwargs.pop('dir', 'Y')
modes = kwargs.pop('modes' , range(D['Nky']))
field = kwargs.pop('field', 'phi')
n_offset = kwargs.pop('offset', 2)
label = kwargs.pop('label', 'ky')
leg_loc = kwargs.pop('loc', 'best')
ncol = kwargs.pop('ncol', 2)
if doCFL == True : pylab.clf()
T = gkcData.getTime(fileh5.root.Analysis.PhaseShift.Time)[2:,1]
if field == 'phi' : n_field = 0
elif field == 'A' : n_field = 1
elif field == 'B' : n_field = 2
else : raise TypeError("Wrong argument for field : " + str(field))
if(dir == 'X'):
pl = plot(T, fileh5.root.Analysis.PhaseShift.X[n_field,:numModes,2:].T)
legend_list = []
for i in range(len(fileh5.root.Analysis.PhaseShift.X[n_field, :numModes,0])):
legend_list.append("kx = %i" % i)
leg = pylab.legend(legend_list, loc='lower right', ncol=2)
leg.draw_frame(0)
elif(dir == 'Y'):
scale = fileh5.root.Grid._v_attrs.Ly/(2. * np.pi)
legend_list = []
for m in modes:
data = fileh5.root.Analysis.PhaseShift.Y[n_field, m,n_offset:]
# set jump value to nan so it is not plotted, (can we speed up using ma ?)
data_m = []
for n in range(len(data[:-1])):
if abs(data[n] - data[n+1]) < 1.: data_m.append(data[n])
else : data_m.append(float('nan'))
data_m.append(data[-1])
data_m = np.array(data_m)
pl = pylab.plot(T, data_m)
if (label == 'm' ) : legend_list.append("m = %i" % m)
elif(label == 'ky') : legend_list.append("ky = %.1f" % (m / scale))
else : print "Name Error"
leg = pylab.legend(legend_list, loc=leg_loc, ncol=ncol, mode="expand").draw_frame(0)
else : raise TypeError("Wrong argument for dir : " + str(dir))
pylab.xlabel("Time")
pylab.xlim((0.,max(T)))
ax = pylab.gca()
ax.set_yticks([-np.pi, -np.pi/2.,0.,np.pi/2., np.pi])
ax.set_yticklabels(['$-\\pi$','$-\\pi/2$','$0$', '$\\pi/2.$', '$\\pi$'])
pylab.ylim((-3.5,3.5))
if field == "phi" : pylab.ylabel("Mode Phase $|\\phi|^2$")
elif field == "A" : pylab.ylabel("Mode Phase $|A_\\parallel|^2$")
elif field == "B" : pylab.ylabel("Mode Phase $|B_\\parallel|^2$")
else : raise TypeError("Wrong argument for field : " + str(field))
return pl, leg
def plotTimeEvolutionModePhase(fileh5, **kwargs):
"""
Plots time evolution of mode power.
Optional keyword arguments:
Keyword Description
=============== ==============================================
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*modes* List of modes (default plot modes).
e.g. modes = [1,4,5] - to plot all modes
modes = range(Nky)[::2] - to plot every second mode
*field* 'phi' electric potential
'A' parallel magnetic vector potential
'B' parallel magnetic field
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*doCFL* clear previous figure
*label* 'ky' or 'm'
*offset* Offset due to zeroset to 2 .
"""
D = gkcData.getDomain(fileh5)
doCFL = kwargs.pop('doCFL', True)
leg_loc = kwargs.pop('loc', 'best')
dir = kwargs.pop('dir', 'Y')
modes = kwargs.pop('modes', range(D['Nky']))
field = kwargs.pop('field', 'phi')
n_offset = kwargs.pop('offset', 2)
label = kwargs.pop('label', 'ky')
leg_loc = kwargs.pop('loc', 'best')
ncol = kwargs.pop('ncol', 2)
if doCFL == True: pylab.clf()
T = gkcData.getTime(fileh5.root.Analysis.PhaseShift.Time)[2:, 1]
if field == 'phi': n_field = 0
elif field == 'A': n_field = 1
elif field == 'B': n_field = 2
else: raise TypeError("Wrong argument for field : " + str(field))
if (dir == 'X'):
pl = plot(T, fileh5.root.Analysis.PhaseShift.X[n_field, :numModes,
2:].T)
legend_list = []
for i in range(
len(fileh5.root.Analysis.PhaseShift.X[n_field, :numModes, 0])):
legend_list.append("kx = %i" % i)
leg = pylab.legend(legend_list, loc='lower right', ncol=2)
leg.draw_frame(0)
elif (dir == 'Y'):
scale = fileh5.root.Grid._v_attrs.Ly / (2. * np.pi)
legend_list = []
for m in modes:
data = fileh5.root.Analysis.PhaseShift.Y[n_field, m, n_offset:]
# set jump value to nan so it is not plotted, (can we speed up using ma ?)
data_m = []
for n in range(len(data[:-1])):
if abs(data[n] - data[n + 1]) < 1.: data_m.append(data[n])
else: data_m.append(float('nan'))
data_m.append(data[-1])
data_m = np.array(data_m)
pl = pylab.plot(T, data_m)
if (label == 'm'): legend_list.append("m = %i" % m)
elif (label == 'ky'): legend_list.append("ky = %.1f" % (m / scale))
else: print "Name Error"
leg = pylab.legend(legend_list, loc=leg_loc, ncol=ncol,
mode="expand").draw_frame(0)
else:
raise TypeError("Wrong argument for dir : " + str(dir))
pylab.xlabel("Time")
pylab.xlim((0., max(T)))
ax = pylab.gca()
ax.set_yticks([-np.pi, -np.pi / 2., 0., np.pi / 2., np.pi])
ax.set_yticklabels(['$-\\pi$', '$-\\pi/2$', '$0$', '$\\pi/2.$', '$\\pi$'])
pylab.ylim((-3.5, 3.5))
if field == "phi": pylab.ylabel("Mode Phase $|\\phi|^2$")
elif field == "A": pylab.ylabel("Mode Phase $|A_\\parallel|^2$")
elif field == "B": pylab.ylabel("Mode Phase $|B_\\parallel|^2$")
else: raise TypeError("Wrong argument for field : " + str(field))
return pl, leg
def plotFrequencyGrowthrates(fileh5, which='b', markline="-", **kwargs):
"""
Plots time evolution of mode power.
Optional keyword arguments:
Keyword Description
=============== ==============================================
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*modes* List of modes (default plot modes).
e.g. modes = [1,4,5] - to plot all modes
modes = range(Nky)[::2] - to plot every second mode
*field* 'phi' electric potential
'A' parallel magnetic vector potential
'B' parallel magnetic field
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*doCFL* clear previous figure
*label* 'ky' or 'm'
*offset* Offset due to zeroset to 2 .
"""
D = gkcData.getDomain(fileh5)
doCLF = kwargs.pop('doCLF', True)
dir = kwargs.pop('dir', 'Y')
modes = kwargs.pop('modes', range(D['Nky']))
field = kwargs.pop('field', 'phi')
n_offset = kwargs.pop('offset', 2)
label = kwargs.pop('label', 'ky')
leg_loc = kwargs.pop('loc', 'best')
start = kwargs.pop('start', 1)
stop = kwargs.pop('stop', -1)
m = kwargs.pop('m', 0)
useLog = kwargs.pop('useLog', True)
if useLog == True: pf = pylab.semilogx
else: pf = pylab.plot
if doCLF == True: pylab.clf()
T = gkcData.getTime(fileh5.root.Analysis.PowerSpectrum.Time)[:, 1]
if field == 'phi': n_field = 0
elif field == 'A': n_field = 1
elif field == 'B': n_field = 2
else: raise TypeError("Wrong argument for field : " + str(field))
if (dir == 'X'):
pl = pf(T, fileh5.root.Analysis.PowerSpectrum.X[n_field, :numModes,
2:].T)
legend_list = []
for i in range(
len(fileh5.root.Analysis.PowerSpectrum.X[n_field, :numModes,
0])):
legend_list.append("kx = %i" % i)
leg = pylab.legend(legend_list, loc='lower right', ncol=2)
leg.draw_frame(0)
elif (dir == 'Y'):
scale = fileh5.root.Grid._v_attrs.Ly / (2. * np.pi)
legend_list = []
if which == 'i' or which == 'b':
power = fileh5.root.Analysis.PowerSpectrum.Y[n_field, :, :]
growthrates = getGrowthrate(T, power, start, stop)
pl = pf(D['ky'], growthrates, "s" + markline, label='$\\gamma$')
if which == 'r' or which == 'b':
shift = fileh5.root.Analysis.PhaseShift.Y[n_field, :, :]
frequency = getFrequency(T, shift, start, stop)
pl = pf(D['ky'], frequency, "v" + markline, label='$\\omega_r$')
#if which !='r' or which != 'i' or which !='b':
# raise TypeError("Wrong argument for which (r/i/b) : " + str(dir))
#pylab.twinx()
pylab.xlim((0.8 * min(D['ky']), 1.2 * max(D['ky'])))
else:
raise TypeError("Wrong argument for dir : " + str(dir))
pylab.xlabel("$k_y$")
leg = pylab.legend(loc=leg_loc, ncol=1, mode="expand").draw_frame(0)
gkcStyle.plotZeroLine(0.8 * min(D['ky']), 1.2 * max(D['ky']))
pylab.ylabel("Growthrate $\\gamma(k_y)$ / Frequency $\\omega_r(k_y)$")
def plotFrequencySpectra(fileh5, which='b', markline="-", **kwargs):
"""
Plots time evolution of mode power.
Optional keyword arguments:
Keyword Description
=============== ==============================================
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*modes* List of modes (default plot modes).
e.g. modes = [1,4,5] - to plot all modes
modes = range(Nky)[::2] - to plot every second mode
*field* 'phi' electric potential
'A' parallel magnetic vector potential
'B' parallel magnetic field
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*doCFL* clear previous figure
*label* 'ky' or 'm'
*offset* Offset due to zeroset to 2 .
"""
import gkcData
import gkcStyle
import pylab
import numpy as np
D = gkcData.getDomain(fileh5)
doCFL = kwargs.pop('doCFL', True)
dir = kwargs.pop('dir', 'Y')
modes = kwargs.pop('modes' , range(D['Nky']))
field = kwargs.pop('field', 'phi')
n_offset = kwargs.pop('offset', 2)
label = kwargs.pop('label', 'ky')
leg_loc = kwargs.pop('loc', 'best')
start = kwargs.pop('start', 1)
stop = kwargs.pop('stop', -1)
if doCFL == True : pylab.clf()
T = gkcData.getTime(fileh5.root.Analysis.PowerSpectrum.Time)[:,1]
if field == 'phi' : n_field = 0
elif field == 'A' : n_field = 1
elif field == 'B' : n_field = 2
else : raise TypeError("Wrong argument for field : " + str(field))
if(dir == 'X'):
raise TypeError("Not implemented for X-direction")
elif(dir == 'Y'):
shift = fileh5.root.Analysis.PhaseShift.Y [n_field, :,start:stop]
print "Using data from T=", T[start], " to T = ", T[stop]
freq = []
for nky in np.arange(1,len(D['ky'])):
time_series = np.sin(shift[nky,:])
FS = np.fft.rfft(time_series)
# Not only if we have constant time step !
freq.append(abs(FS))
fftfreq = np.fft.fftfreq(len(abs(np.fft.fftshift(FS))), d = (T[-10]-T[-11]))
freq = np.array(freq)
print np.shape(fftfreq), " 2 : ", np.shape(D['ky'][1:]), np.shape(freq)
pylab.contourf(D['ky'][1:], np.fft.fftshift(fftfreq), freq.T, 100, cmap=pylab.cm.jet)
pylab.xlim((D['ky'][1], D['ky'][-1]))
pylab.ylim((min(fftfreq), max(fftfreq)))
pylab.colorbar()
else : raise TypeError("Wrong argument for dir : " + str(dir))
pylab.gca().set_xscale("log")
pylab.xlabel("$k_y$")
gkcStyle.plotZeroLine(D['ky'][1], D['ky'][-1], color='r')
pylab.ylabel("Frequency $\\omega_r(k_y)$")
results = []
##################### Call for adiabatic case ######################
# Open Files and extract frequency and growthrates
fileh5 = tables.openFile(sys.argv[1])
isKinetic = (len(fileh5.root.Species.cols[:]) == 3)
rhoLn = fileh5.root.Species.cols.w_n[1]
rhoLT = fileh5.root.Species.cols.w_T[1]
#kp = fileh5.root.Species.cols.w_T[1]
D = gkcData.getDomain(fileh5)
omega_sim = gkcLinear.getFrequencyGrowthrates(fileh5, start=50, stop=-1)
# get Theory
ky_list = np.logspace(np.log10(D['ky'][1]), np.log10(D['ky'][-1]), 256)
if (isKinetic):
mode = 'EM'
m_ie = fileh5.root.Species.cols.Mass[1] / fileh5.root.Species.cols.Mass[2]
else:
mode = 'ETG'
m_ie = 1.
param = { 'disp' : 'GyroSlab', 'mode' : mode, 'beta' : 1.0e-5, 'tau' : 1., 'lambdaD' : 1.e-5, 'rhoLn' : rhoLn, \
'rhoLT' : rhoLT, 'kx' : np.sqrt(2.) * 1.e-1 , 'kp' : 2. * np.sqrt(2.) * 1.e-3, 'm_ie' : m_ie, 'adiab' : lambda x : 1 }
##################### Call for adiabatic case ######################
# Open Files and extract frequency and growthrates
fileh5 = tables.openFile(sys.argv[1])
isKinetic = (len(fileh5.root.Species.cols[:]) == 3)
rhoLn = fileh5.root.Species.cols.w_n[1]
rhoLT = fileh5.root.Species.cols.w_T[1]
#kp = fileh5.root.Species.cols.w_T[1]
D = gkcData.getDomain(fileh5)
omega_sim = gkcLinear.getFrequencyGrowthrates(fileh5, start=50, stop=-1)
# get Theory
ky_list = np.logspace(np.log10(D['ky'][1]), np.log10(D['ky'][-1]), 256)
if (isKinetic) :
mode = 'EM'
m_ie = fileh5.root.Species.cols.Mass[1]/fileh5.root.Species.cols.Mass[2]
else :
mode = 'ETG'
m_ie = 1.
param = { 'disp' : 'GyroSlab', 'mode' : mode, 'beta' : 1.0e-5, 'tau' : 1., 'lambdaD' : 1.e-5, 'rhoLn' : rhoLn, \
'rhoLT' : rhoLT, 'kx' : np.sqrt(2.) * 1.e-1 , 'kp' : 2. * np.sqrt(2.) * 1.e-3, 'm_ie' : m_ie, 'adiab' : lambda x : 1 }
def plotModeStructure(fileh5, mode, part = "b", phaseCorrect=True, **kwargs):
"""
Plots time evolution of mode power.
Optional keyword arguments:
Keyword Description
=============== ==============================================
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*modes* List of modes (default plot modes).
e.g. modes = [1,4,5] - to plot all modes
modes = range(Nky)[::2] - to plot every second mode
*field* 'phi' electric potential
'A' parallel magnetic vector potential
'B' parallel magnetic field
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*doCFL* clear previous figure
*label* 'ky' or 'm'
*offset* Offset due to zeroset to 2 .
"""
D = gkcData.getDomain(fileh5)
doCFL = kwargs.pop('doCFL', True)
field = kwargs.pop('field', 'phi')
Z = kwargs.pop('Z', 0)
frame = kwargs.pop('frame', -1)
label = kwargs.pop('label', "")
norm = kwargs.pop('norm', 'max')
m = kwargs.pop('m', 0)
phase = kwargs.pop('phase', 0.)
color = kwargs.pop('color', '')
phase = np.exp(1.j * phase)
if field == 'phi' :
data_X = fileh5.root.Visualization.Phi[Z,mode,:,frame]
y_label = '$\\phi(x)$'
elif field == 'A' :
data_X = fileh5.root.Visualization.Ap[Z,mode,:,frame]
n_field = 2
y_label = '$A_\parallel(x)$'
elif field == 'B' :
n_field = 2
data_X = fileh5.root.Visualization.Bp[Z,mode,:,frame]
y_label = '$B_\\parallel(x)$'
else : raise TypeError("Wrong argument for field : " + str(field))
print "Mode Structure at T = ", gkcData.getTime(fileh5.root.Visualization.Time)[frame,:]
# Normalization
def Normalize_data(data):
print norm
if norm == 'sum2' : return np.sum(abs(data))
elif norm == 'max' : return abs(data).max()
else : raise TypeError("No such normalization")
if(phaseCorrect == True):
# Calculate phase
phase_shift = np.arctan2(np.sum(np.imag(np.sum(data_X))), np.real(np.sum(data_X)))
print "phaseCorrect = True : Correcting for phase ", phase_shift
data_X = data_X * np.exp(- 1.j * phase_shift)
# Normalization is to real part
if part == "a" :
if color=='' : color='g'
data_X = abs(data_X) / Normalize_data(data_X)
pylab.plot(D['X'], data_X, color=color, label=label)
if part in [ 'r' , 'b' ]:
if color=='' : color='r'
data_X = np.real(data_X * phase) / Normalize_data(np.real(data_X))
pylab.plot(D['X'], data_X, 'o-', color=color, label=label, markersize=8., markeredgecolor='None')
if part in [ 'i', 'b' ] :
if color=='' : color=gkcStyle.color_indigo
data_X = np.imag(data_X * phase) / Normalize_data(np.real(data_X))
pylab.plot(D['X'], data_X, 's-', color=color, label=label, markersize=8., markeredgecolor='None')
#else : raise TypeError("Wrong argument for part : " + str(part))
if field == "phi" : pylab.ylabel("Mode Power $|\\phi|^2$")
elif field == "A" : pylab.ylabel("Mode Power $|A_\\parallel|^2$")
elif field == "B" : pylab.ylabel("Mode Power $|B_\\parallel|^2$")
else : raise TypeError("Wrong argument for field : " + str(field))
pylab.xlim((min(D['X']), max(D['X'])))
pylab.xlabel("X")
pylab.ylabel(y_label)
def plotModeStructure(fileh5, mode, part="b", phaseCorrect=True, **kwargs):
"""
Plots time evolution of mode power.
Optional keyword arguments:
Keyword Description
=============== ==============================================
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*modes* List of modes (default plot modes).
e.g. modes = [1,4,5] - to plot all modes
modes = range(Nky)[::2] - to plot every second mode
*field* 'phi' electric potential
'A' parallel magnetic vector potential
'B' parallel magnetic field
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*doCFL* clear previous figure
*label* 'ky' or 'm'
*offset* Offset due to zeroset to 2 .
"""
D = gkcData.getDomain(fileh5)
doCFL = kwargs.pop('doCFL', True)
field = kwargs.pop('field', 'phi')
Z = kwargs.pop('Z', 0)
frame = kwargs.pop('frame', -1)
label = kwargs.pop('label', "")
norm = kwargs.pop('norm', 'max')
m = kwargs.pop('m', 0)
phase = kwargs.pop('phase', 0.)
color = kwargs.pop('color', '')
phase = np.exp(1.j * phase)
if field == 'phi':
data_X = fileh5.root.Visualization.Phi[Z, mode, :, frame]
y_label = '$\\phi(x)$'
elif field == 'A':
data_X = fileh5.root.Visualization.Ap[Z, mode, :, frame]
n_field = 2
y_label = '$A_\parallel(x)$'
elif field == 'B':
n_field = 2
data_X = fileh5.root.Visualization.Bp[Z, mode, :, frame]
y_label = '$B_\\parallel(x)$'
else:
raise TypeError("Wrong argument for field : " + str(field))
print "Mode Structure at T = ", gkcData.getTime(
fileh5.root.Visualization.Time)[frame, :]
# Normalization
def Normalize_data(data):
print norm
if norm == 'sum2': return np.sum(abs(data))
elif norm == 'max': return abs(data).max()
else: raise TypeError("No such normalization")
if (phaseCorrect == True):
# Calculate phase
phase_shift = np.arctan2(np.sum(np.imag(np.sum(data_X))),
np.real(np.sum(data_X)))
print "phaseCorrect = True : Correcting for phase ", phase_shift
data_X = data_X * np.exp(-1.j * phase_shift)
# Normalization is to real part
if part == "a":
if color == '': color = 'g'
data_X = abs(data_X) / Normalize_data(data_X)
pylab.plot(D['X'], data_X, color=color, label=label)
if part in ['r', 'b']:
if color == '': color = 'r'
data_X = np.real(data_X * phase) / Normalize_data(np.real(data_X))
pylab.plot(D['X'],
data_X,
'o-',
color=color,
label=label,
markersize=8.,
markeredgecolor='None')
if part in ['i', 'b']:
if color == '': color = gkcStyle.color_indigo
data_X = np.imag(data_X * phase) / Normalize_data(np.real(data_X))
pylab.plot(D['X'],
data_X,
's-',
color=color,
label=label,
markersize=8.,
markeredgecolor='None')
#else : raise TypeError("Wrong argument for part : " + str(part))
if field == "phi": pylab.ylabel("Mode Power $|\\phi|^2$")
elif field == "A": pylab.ylabel("Mode Power $|A_\\parallel|^2$")
elif field == "B": pylab.ylabel("Mode Power $|B_\\parallel|^2$")
else: raise TypeError("Wrong argument for field : " + str(field))
pylab.xlim((min(D['X']), max(D['X'])))
pylab.xlabel("X")
pylab.ylabel(y_label)
def plotTimeEvolutionModePower(fileh5, **kwargs):
"""
Plots time evolution of mode power.
Optional keyword arguments:
Keyword Description
=============== ==============================================
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*modes* List of modes (default plot modes).
e.g. modes = [1,4,5] - to plot all modes
modes = range(Nky)[::2] - to plot every second mode
*field* 'phi' electric potential
'A' parallel magnetic vector potential
'B' parallel magnetic field
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*doCFL* clear previous figure
*label* 'ky' or 'm'
*offset* Offset due to zeroset to 2 .
"""
D = gkcData.getDomain(fileh5)
doCFL = kwargs.pop('doCFL', True)
doLog = kwargs.pop('doLog', True)
dir = kwargs.pop('dir', 'Y')
modes = kwargs.pop('modes' , range(D['Nky']))
field = kwargs.pop('field', 'phi')
n_offset = kwargs.pop('offset', 3)
label = kwargs.pop('label', 'k')
leg_loc = kwargs.pop('loc', 'best')
ncol = kwargs.pop('ncol', 3)
showLegend = kwargs.pop('showLegend', True)
if doCFL == True : pylab.clf()
T = gkcData.getTime(fileh5.root.Analysis.PowerSpectrum.Time)[:,1]
if field == 'phi' : n_field = 0
elif field == 'A' : n_field = 1
elif field == 'B' : n_field = 2
else : raise TypeError("Wrong argument for field : " + str(field))
legend_list = []
if(dir == 'X'):
for n in modes:
pl = pylab.semilogy(T[n_offset:], fileh5.root.Analysis.PowerSpectrum.X[n_field, n,n_offset:].T)
if (label == 'm' ) : legend_list.append("n = %i" % n)
elif(label == 'k') : legend_list.append("kx = %.1f" % (D['kx'][n]))
else : print "Name Error"
elif(dir == 'Y'):
legend_list = []
for m in modes:
if(doLog) : pylab.semilogy(T[n_offset:], fileh5.root.Analysis.PowerSpectrum.Y[n_field, m,n_offset:].T, label='$k_y^{(%i)} = %.2f$' % (m, D['ky'][m]))
else : pylab.plot(T[n_offset:], fileh5.root.Analysis.PowerSpectrum.Y[n_field, m,n_offset:].T, label='$k_y^{(%i)} = %.2f$' % (m, D['ky'][m]))
else : raise TypeError("Wrong argument for dir : " + str(dir))
#if showLegend == True:
leg = pylab.legend(loc=leg_loc, ncol=ncol, mode="expand").draw_frame(0)
pylab.xlabel("Time")
pylab.xlim((0.,max(T)))
if field == "phi" : pylab.ylabel("Mode Power $|\\phi|^2$")
elif field == "A" : pylab.ylabel("Mode Power $|A_\\parallel|^2$")
elif field == "B" : pylab.ylabel("Mode Power $|B_\\parallel|^2$")
else : raise TypeError("Wrong argument for field : " + str(field))
def plotInstantGrowthrates(fileh5, **kwargs):
"""
Plots instant growthrates of mode power
Optional keyword arguments:
Keyword Description
=============== ==============================================
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*modes* List of modes (default plot modes).
e.g. modes = [1,4,5] - to plot all modes
modes = range(Nky)[::2] - to plot every second mode
*field* 'phi' electric potential
'A' parallel magnetic vector potential
'B' parallel magnetic field
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*doCFL* clear previous figure
*label* 'ky' or 'm'
*offset* Offset due to zeroset to 2 .
"""
import scipy.ndimage
import scipy.interpolate
D = gkcData.getDomain(fileh5)
doCFL = kwargs.pop('doCFL', True)
dir = kwargs.pop('dir', 'Y')
modes = kwargs.pop('modes' , range(D['Nky']))
field = kwargs.pop('field', 'phi')
n_offset = kwargs.pop('offset', 2)
label = kwargs.pop('label', 'ky')
leg_loc = kwargs.pop('loc', 'best')
off = kwargs.pop('off', 2)
sigma = kwargs.pop('sigma', 10)
#filterType = kwargs.pop('filterType', 'hanning')
if doCFL == True : pylab.clf()
T = gkcData.getTime(fileh5.root.Analysis.PowerSpectrum.Time)[:,1]
# TimeStep is irregular thus needs to cast into regular
def cast2Equidistant(T, Var):
f = scipy.interpolate.interp1d(T, Var)
Tnew = np.linspace(min(T), max(T), 1000)
return Tnew, f(Tnew)
if field == 'phi' : n_field = 0
elif field == 'A' : n_field = 1
elif field == 'B' : n_field = 2
else : raise TypeError("Wrong argument for field : " + str(field))
if(dir == 'X'):
plot(T, grad)
legend_list = []
for i in range(len(fileh5.root.Analysis.PowerSpectrum.X[n_field,:numModes,0])):
legend_list.append("kx = %i" % i)
leg = legend(legend_list, loc='best', ncol=3)
leg.draw_frame(0)
if(dir == 'Y'):
for m in modes:
Var = fileh5.root.Analysis.PowerSpectrum.Y[n_field,m,off:]
Tn, V = cast2Equidistant(T[off:], np.log(Var))
# Use NdImage for line-smoothening (convolution with gaussian kernel)
V = scipy.ndimage.gaussian_filter(V, sigma=sigma, mode='nearest')
gamma = np.gradient(V, Tn[1]-Tn[0])
pylab.plot(Tn, gamma, "-", label='$k_y^{(%i)} = %.2f$' % (m, D['ky'][m]))#, color=gkcStyle.markers_C[m])
leg = pylab.legend(loc='best', ncol=3).draw_frame(0)
pylab.xlabel("Time")
pylab.xlim((0.,max(T)))
pylab.ylabel("Instant Mode Growth $\gamma$")
def plotFrequencyGrowthrates(fileh5, which='b', markline="-", **kwargs):
"""
Plots time evolution of mode power.
Optional keyword arguments:
Keyword Description
=============== ==============================================
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*modes* List of modes (default plot modes).
e.g. modes = [1,4,5] - to plot all modes
modes = range(Nky)[::2] - to plot every second mode
*field* 'phi' electric potential
'A' parallel magnetic vector potential
'B' parallel magnetic field
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*doCFL* clear previous figure
*label* 'ky' or 'm'
*offset* Offset due to zeroset to 2 .
"""
D = gkcData.getDomain(fileh5)
doCFL = kwargs.pop('doCFL', True)
dir = kwargs.pop('dir', 'Y')
modes = kwargs.pop('modes' , range(D['Nky']))
field = kwargs.pop('field', 'phi')
n_offset = kwargs.pop('offset', 2)
label = kwargs.pop('label', 'ky')
leg_loc = kwargs.pop('loc', 'best')
start = kwargs.pop('start', 1)
stop = kwargs.pop('stop', -1)
if doCFL == True : pylab.clf()
T = gkcData.getTime(fileh5.root.Analysis.PowerSpectrum.Time)[:,1]
print "Fitting from T : ", T[start], " - ", T[stop]
if field == 'phi' : n_field = 0
elif field == 'A' : n_field = 1
elif field == 'B' : n_field = 2
else : raise TypeError("Wrong argument for field : " + str(field))
if(dir == 'X'):
pl = semilogy(T, fileh5.root.Analysis.PowerSpectrum.X[n_field,:numModes,2:].T)
legend_list = []
for i in range(len(fileh5.root.Analysis.PowerSpectrum.X[n_field, :numModes,0])):
legend_list.append("kx = %i" % i)
leg = pylab.legend(legend_list, loc='lower right', ncol=2)
leg.draw_frame(0)
elif(dir == 'Y'):
scale = fileh5.root.Grid._v_attrs.Ly/(2. * np.pi)
legend_list = []
if which=='i' or which=='b':
power = fileh5.root.Analysis.PowerSpectrum.Y[n_field, :,:]
growthrates = getGrowthrate(T,power, start,stop, dir='Y')
pl = pylab.semilogx(D['ky'], growthrates, "s" + markline, label='$\\gamma$')
if which=='r' or which=='b':
shift = fileh5.root.Analysis.PhaseShift.Y [n_field, :,:]
frequency = getFrequency(T,shift, start,stop, dir='Y')
pl = pylab.semilogx(D['ky'], frequency, "v" + markline, label='$\\omega_r$')
#if which !='r' or which != 'i' or which !='b':
# raise TypeError("Wrong argument for which (r/i/b) : " + str(dir))
#pylab.twinx()
pylab.xlim((0.8*min(D['ky']), 1.2*max(D['ky'])))
else : raise TypeError("Wrong argument for dir : " + str(dir))
pylab.xlabel("$k_y$")
leg = pylab.legend(loc=leg_loc, ncol=1, mode="expand").draw_frame(0)
gkcStyle.plotZeroLine(0.8*min(D['ky']), 1.2*max(D['ky']))
pylab.ylabel("Growthrate $\\gamma(k_y)$ / Frequency $\\omega_r(k_y)$")
def plotFrequencySpectra(fileh5, which='b', markline="-", **kwargs):
"""
Plots time evolution of mode power.
Optional keyword arguments:
Keyword Description
=============== ==============================================
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*modes* List of modes (default plot modes).
e.g. modes = [1,4,5] - to plot all modes
modes = range(Nky)[::2] - to plot every second mode
*field* 'phi' electric potential
'A' parallel magnetic vector potential
'B' parallel magnetic field
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*doCFL* clear previous figure
*label* 'ky' or 'm'
*offset* Offset due to zeroset to 2 .
"""
import gkcData
import gkcStyle
import pylab
import numpy as np
D = gkcData.getDomain(fileh5)
doCFL = kwargs.pop('doCFL', True)
dir = kwargs.pop('dir', 'Y')
modes = kwargs.pop('modes', range(D['Nky']))
field = kwargs.pop('field', 'phi')
n_offset = kwargs.pop('offset', 2)
label = kwargs.pop('label', 'ky')
leg_loc = kwargs.pop('loc', 'best')
start = kwargs.pop('start', 1)
stop = kwargs.pop('stop', -1)
if doCFL == True: pylab.clf()
T = gkcData.getTime(fileh5.root.Analysis.PowerSpectrum.Time)[:, 1]
if field == 'phi': n_field = 0
elif field == 'A': n_field = 1
elif field == 'B': n_field = 2
else: raise TypeError("Wrong argument for field : " + str(field))
if (dir == 'X'):
raise TypeError("Not implemented for X-direction")
elif (dir == 'Y'):
shift = fileh5.root.Analysis.PhaseShift.Y[n_field, :, start:stop]
print "Using data from T=", T[start], " to T = ", T[stop]
freq = []
for nky in np.arange(1, len(D['ky'])):
time_series = np.sin(shift[nky, :])
FS = np.fft.rfft(time_series)
# Not only if we have constant time step !
freq.append(abs(FS))
fftfreq = np.fft.fftfreq(len(abs(np.fft.fftshift(FS))),
d=(T[-10] - T[-11]))
freq = np.array(freq)
print np.shape(fftfreq), " 2 : ", np.shape(D['ky'][1:]), np.shape(freq)
pylab.contourf(D['ky'][1:],
np.fft.fftshift(fftfreq),
freq.T,
100,
cmap=pylab.cm.jet)
pylab.xlim((D['ky'][1], D['ky'][-1]))
pylab.ylim((min(fftfreq), max(fftfreq)))
pylab.colorbar()
else:
raise TypeError("Wrong argument for dir : " + str(dir))
pylab.gca().set_xscale("log")
pylab.xlabel("$k_y$")
gkcStyle.plotZeroLine(D['ky'][1], D['ky'][-1], color='r')
pylab.ylabel("Frequency $\\omega_r(k_y)$")
def plotAveragedFluxes(fileh5, var='H', **kwargs):
"""
Plots time evolution of mode power.
Optional keyword arguments:
Keyword Description
=============== ==============================================
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*modes* List of modes (default plot modes).
e.g. modes = [1,4,5] - to plot all modes
modes = range(Nky)[::2] - to plot every second mode
*field* 'phi' electric potential
'A' parallel magnetic vector potential
'B' parallel magnetic field
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*doCFL* clear previous figure
*label* 'ky' or 'm'
*offset* Offset due to zeroset to 2 .
"""
import gkcData
import gkcStyle
import pylab
import numpy as np
D = gkcData.getDomain(fileh5)
doCFL = kwargs.pop('doCFL', True)
start = kwargs.pop('start', 1)
stop = kwargs.pop('stop', -1)
scale = kwargs.pop('scale', 'ky')
field = kwargs.pop('field', 'phi')
if doCFL == True : pylab.clf()
if field == 'phi' : n_field = 0
elif field == 'A' : n_field = 1
elif field == 'B' : n_field = 2
else : raise TypeError("Wrong argument for field : " + str(field))
T = gkcData.getTime(fileh5.root.Analysis.PowerSpectrum.Time)[:,1]
# Averaged over Z and start stop
if var == 'P' :
data = np.mean(fileh5.root.Analysis.Flux.Density[n_field,:,:,start:stop], axis=2)
ylabel = "Particle Flux $k_y \\Gamma/\\Gamma_{gB}\,(ky)$"
elif var == 'H' :
data = np.mean(fileh5.root.Analysis.Flux.Heat [n_field,:,:,start:stop], axis=2)
#ylabel = "Heat Flux $Q/Q_\\textrm{gB}(k_y)$"
ylabel = "Heat Flux $k_y Q/Q_{gB}\,(k_y)$"
else : raise TypeError("No such variable")
if scale == 'ky' : gor_ky = D['ky'][1:]
else : gor_ky = 1.
for s in range(D['Ns']):
species_name = fileh5.root.Species.cols.Name[s+1]
pylab.semilogx(D['ky'][1:], gor_ky * data[1:,s], gkcStyle.markers_D[s]+'-', label=species_name, color=gkcStyle.markers_C[s], markersize=7.)
pylab.xlabel("$k_y \\rho_i$")
pylab.ylabel(ylabel)
pylab.legend(ncol=2).draw_frame(0)
pylab.xlim((0.8*D['ky'][1], 1.2*D['ky'][-1]))
def plotInstantGrowthrates(fileh5, **kwargs):
"""
Plots instant growthrates of mode power
Optional keyword arguments:
Keyword Description
=============== ==============================================
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*modes* List of modes (default plot modes).
e.g. modes = [1,4,5] - to plot all modes
modes = range(Nky)[::2] - to plot every second mode
*field* 'phi' electric potential
'A' parallel magnetic vector potential
'B' parallel magnetic field
*dir* Direction 'X' (for radial) or 'Y' for poloidal
*doCFL* clear previous figure
*label* 'ky' or 'm'
*offset* Offset due to zeroset to 2 .
"""
import scipy.ndimage
import scipy.interpolate
D = gkcData.getDomain(fileh5)
doCFL = kwargs.pop('doCFL', True)
dir = kwargs.pop('dir', 'Y')
modes = kwargs.pop('modes', range(D['Nky']))
field = kwargs.pop('field', 'phi')
n_offset = kwargs.pop('offset', 2)
label = kwargs.pop('label', 'ky')
leg_loc = kwargs.pop('loc', 'best')
off = kwargs.pop('off', 2)
sigma = kwargs.pop('sigma', 10)
#filterType = kwargs.pop('filterType', 'hanning')
if doCFL == True: pylab.clf()
T = gkcData.getTime(fileh5.root.Analysis.PowerSpectrum.Time)[:, 1]
# TimeStep is irregular thus needs to cast into regular
def cast2Equidistant(T, Var):
f = scipy.interpolate.interp1d(T, Var)
Tnew = np.linspace(min(T), max(T), 1000)
return Tnew, f(Tnew)
if field == 'phi': n_field = 0
elif field == 'A': n_field = 1
elif field == 'B': n_field = 2
else: raise TypeError("Wrong argument for field : " + str(field))
if (dir == 'X'):
plot(T, grad)
legend_list = []
for i in range(
len(fileh5.root.Analysis.PowerSpectrum.X[n_field, :numModes,
0])):
legend_list.append("kx = %i" % i)
leg = legend(legend_list, loc='best', ncol=3)
leg.draw_frame(0)
if (dir == 'Y'):
for m in modes:
Var = fileh5.root.Analysis.PowerSpectrum.Y[n_field, m, off:]
Tn, V = cast2Equidistant(T[off:], np.log(Var))
# Use NdImage for line-smoothening (convolution with gaussian kernel)
V = scipy.ndimage.gaussian_filter(V, sigma=sigma, mode='nearest')
gamma = np.gradient(V, Tn[1] - Tn[0])
pylab.plot(Tn,
gamma,
"-",
label='$k_y^{(%i)} = %.2f$' %
(m, D['ky'][m])) #, color=gkcStyle.markers_C[m])
leg = pylab.legend(loc='best', ncol=3).draw_frame(0)
pylab.xlabel("Time")
pylab.xlim((0., max(T)))
pylab.ylabel("Instant Mode Growth $\gamma$")
